When deciding to order a diagnostic test, how should pretest probability be used?

• A. Use pretest probability to determine impact of a test via likelihood ratios.
• B. Ignore pretest probability and rely only on test characteristics.
• C. Set pretest probability to 50% in all cases.
• D. Only test when pretest probability is 0%.

Answer: (A)

Pretest probability is how likely you think the disease is before testing, and its real value is in shaping how informative a test will be for a given patient. The test’s usefulness isn’t determined by sensitivity and specificity alone; what matters is how much the result changes your estimate of disease. That change is quantified by likelihood ratios. A positive result increases your odds by the positive likelihood ratio, while a negative result decreases them by the negative likelihood ratio. By converting the pretest probability to pretest odds, multiplying by the appropriate likelihood ratio, and converting back to a probability, you get the post-test probability. This post-test probability guides management—whether to treat, observe, or pursue further testing—based on thresholds for action. In short, use pretest probability to gauge how much a test will move your diagnostic belief, not just to know the test’s characteristics in isolation.